Add logic for parsing references from last page of PDF

src/main/java/org/jabref/gui/actions/SimpleCommand.java

@@ -15,10 +15,6 @@ public abstract class SimpleCommand extends CommandBase {
 
     protected ReadOnlyStringWrapper statusMessage = new ReadOnlyStringWrapper("");
 
-    public String getStatusMessage() {
-        return statusMessage.get();
-    }
-
     public ReadOnlyStringProperty statusMessageProperty() {
         return statusMessage.getReadOnlyProperty();
     }

src/main/java/org/jabref/gui/maintable/ExtractReferencesAction.java

@@ -1,9 +1,13 @@
 package org.jabref.gui.maintable;
 
 import java.nio.file.Path;
-import java.util.LinkedList;
+import java.util.Iterator;
 import java.util.List;
+import java.util.Optional;
+import java.util.StringJoiner;
 import java.util.concurrent.Callable;
+import java.util.regex.Matcher;
+import java.util.regex.Pattern;
 
 import org.jabref.gui.DialogService;
 import org.jabref.gui.StateManager;
@@ -13,13 +17,22 @@
 import org.jabref.gui.util.BackgroundTask;
 import org.jabref.gui.util.TaskExecutor;
 import org.jabref.logic.importer.ParserResult;
+import org.jabref.logic.importer.fileformat.BibliopgraphyFromPdfImporter;
 import org.jabref.logic.importer.util.GrobidService;
 import org.jabref.logic.l10n.Localization;
 import org.jabref.logic.util.io.FileUtil;
+import org.jabref.model.database.BibDatabaseContext;
 import org.jabref.model.entry.BibEntry;
 import org.jabref.model.entry.LinkedFile;
+import org.jabref.model.entry.field.StandardField;
 import org.jabref.preferences.PreferencesService;
 
+import org.jspecify.annotations.NonNull;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * SIDE EFFECT: Sets the "cites" field of the entry having the linked files
+ */
 public class ExtractReferencesAction extends SimpleCommand {
     private final int FILES_LIMIT = 10;
 
@@ -29,6 +42,7 @@ public class ExtractReferencesAction extends SimpleCommand {
     private final BibEntry entry;
     private final LinkedFile linkedFile;
     private final TaskExecutor taskExecutor;
+    private final BibliopgraphyFromPdfImporter bibliopgraphyFromPdfImporter;
 
     public ExtractReferencesAction(DialogService dialogService,
                                    StateManager stateManager,
@@ -37,24 +51,30 @@ public ExtractReferencesAction(DialogService dialogService,
         this(dialogService, stateManager, preferencesService, null, null, taskExecutor);
     }
 
-    public ExtractReferencesAction(DialogService dialogService,
-                                   StateManager stateManager,
-                                   PreferencesService preferencesService,
-                                   BibEntry entry,
-                                   LinkedFile linkedFile,
-                                   TaskExecutor taskExecutor) {
+    /**
+     * Can be used to bind the action on a context menu in the linked file view (future work)
+     *
+     * @param entry the entry to handle (can be null)
+     * @param linkedFile the linked file (can be null)
+     */
+    private ExtractReferencesAction(@NonNull DialogService dialogService,
+                                    @NonNull StateManager stateManager,
+                                    @NonNull PreferencesService preferencesService,
+                                    @Nullable BibEntry entry,
+                                    @Nullable LinkedFile linkedFile,
+                                    @NonNull TaskExecutor taskExecutor) {
         this.dialogService = dialogService;
         this.stateManager = stateManager;
         this.preferencesService = preferencesService;
         this.entry = entry;
         this.linkedFile = linkedFile;
         this.taskExecutor = taskExecutor;
+        bibliopgraphyFromPdfImporter = new BibliopgraphyFromPdfImporter(preferencesService.getCitationKeyPatternPreferences());
 
         if (this.linkedFile == null) {
             this.executable.bind(
                     ActionHelper.needsEntriesSelected(stateManager)
                                 .and(ActionHelper.hasLinkedFileForSelectedEntries(stateManager))
-                                .and(this.preferencesService.getGrobidPreferences().grobidEnabledProperty())
             );
         } else {
             this.setExecutable(true);
@@ -68,34 +88,113 @@ public void execute() {
 
     private void extractReferences() {
         stateManager.getActiveDatabase().ifPresent(databaseContext -> {
-            List<BibEntry> selectedEntries = new LinkedList<>();
+            boolean online = this.preferencesService.getGrobidPreferences().isGrobidEnabled();
+
+            List<BibEntry> selectedEntries;
             if (entry == null) {
                 selectedEntries = stateManager.getSelectedEntries();
             } else {
-                selectedEntries.add(entry);
+                selectedEntries = List.of(entry);
             }
 
-            List<Path> fileList = FileUtil.getListOfLinkedFiles(selectedEntries, databaseContext.getFileDirectories(preferencesService.getFilePreferences()));
-            if (fileList.size() > FILES_LIMIT) {
-                boolean continueOpening = dialogService.showConfirmationDialogAndWait(Localization.lang("Processing a large number of files"),
-                        Localization.lang("You are about to process %0 files. Continue?", fileList.size()),
-                        Localization.lang("Continue"), Localization.lang("Cancel"));
-                if (!continueOpening) {
+            Callable<ParserResult> parserResultCallable;
+            if (online) {
+                parserResultCallable = getParserResultCallableOnline(databaseContext, selectedEntries);
+                if (parserResultCallable == null) {
                     return;
                 }
+            } else {
+                parserResultCallable = getParserResultCallableOffline(databaseContext, selectedEntries);
             }
-
-            Callable<ParserResult> parserResultCallable = () -> new ParserResult(
-                    new GrobidService(this.preferencesService.getGrobidPreferences()).processReferences(fileList, preferencesService.getImportFormatPreferences())
-            );
             BackgroundTask<ParserResult> task = BackgroundTask.wrap(parserResultCallable)
                                                               .withInitialMessage(Localization.lang("Processing PDF(s)"));
 
             task.onFailure(dialogService::showErrorDialogAndWait);
 
             ImportEntriesDialog dialog = new ImportEntriesDialog(stateManager.getActiveDatabase().get(), task);
-            dialog.setTitle(Localization.lang("Extract References"));
+            String title;
+            if (online) {
+                title = Localization.lang("Extract References (online)");
+            } else {
+                title = Localization.lang("Extract References (offline)");
+            }
+            dialog.setTitle(title);
             dialogService.showCustomDialogAndWait(dialog);
         });
     }
+
+    private @NonNull Callable<ParserResult> getParserResultCallableOffline(BibDatabaseContext databaseContext, List<BibEntry> selectedEntries) {
+        return () -> {
+            BibEntry currentEntry = selectedEntries.getFirst();
+            List<Path> fileList = FileUtil.getListOfLinkedFiles(selectedEntries, databaseContext.getFileDirectories(preferencesService.getFilePreferences()));
+
+            // We need to have ParserResult handled at the importer, because it imports the meta data (library type, encoding, ...)
+            ParserResult result = bibliopgraphyFromPdfImporter.importDatabase(fileList.getFirst());
+
+            // subsequent files are just appended to result
+            Iterator<Path> fileListIterator = fileList.iterator();
+            fileListIterator.next(); // skip first file
+            extractReferences(fileListIterator, result, currentEntry);
+
+            // handle subsequent entries
+            Iterator<BibEntry> selectedEntriesIterator = selectedEntries.iterator();
+            selectedEntriesIterator.next(); // skip first entry
+            while (selectedEntriesIterator.hasNext()) {
+                currentEntry = selectedEntriesIterator.next();
+                fileList = FileUtil.getListOfLinkedFiles(List.of(currentEntry), databaseContext.getFileDirectories(preferencesService.getFilePreferences()));
+                fileListIterator = fileList.iterator();
+                extractReferences(fileListIterator, result, currentEntry);
+            }
+
+            return result;
+        };
+    }
+
+    private void extractReferences(Iterator<Path> fileListIterator, ParserResult result, BibEntry currentEntry) {
+        while (fileListIterator.hasNext()) {
+            result.getDatabase().insertEntries(bibliopgraphyFromPdfImporter.importDatabase(fileListIterator.next()).getDatabase().getEntries());
+        }
+
+        StringJoiner cites = new StringJoiner(",");
+        int count = 0;
+        for (BibEntry importedEntry : result.getDatabase().getEntries()) {
+            count++;
+            Optional<String> citationKey = importedEntry.getCitationKey();
+            if (citationKey.isPresent()) {
+                cites.add(citationKey.get());
+            } else {
+                String sourceCitationKey = currentEntry.getCitationKey().orElse("unknown");
+                String newCitationKey;
+                // Could happen if no author and no year is present
+                // We use the number of the comment field (because there is no other way to get the number reliable)
+                Pattern pattern = Pattern.compile("^\\[(\\d+)\\]");
+                Matcher matcher = pattern.matcher(importedEntry.getField(StandardField.COMMENT).orElse(""));
+                if (matcher.hasMatch()) {
+                    newCitationKey = sourceCitationKey + "-" + matcher.group(1);
+                } else {
+                    newCitationKey = sourceCitationKey + "-" + count;
+                }
+                importedEntry.setCitationKey(newCitationKey);
+                cites.add(newCitationKey);
+            }
+        }
+        currentEntry.setField(StandardField.CITES, cites.toString());
+    }
+
+    private @Nullable Callable<ParserResult> getParserResultCallableOnline(BibDatabaseContext databaseContext, List<BibEntry> selectedEntries) {
+        Callable<ParserResult> parserResultCallable;
+        List<Path> fileList = FileUtil.getListOfLinkedFiles(selectedEntries, databaseContext.getFileDirectories(preferencesService.getFilePreferences()));
+        if (fileList.size() > FILES_LIMIT) {
+            boolean continueOpening = dialogService.showConfirmationDialogAndWait(Localization.lang("Processing a large number of files"),
+                    Localization.lang("You are about to process %0 files. Continue?", fileList.size()),
+                    Localization.lang("Continue"), Localization.lang("Cancel"));
+            if (!continueOpening) {
+                return null;
+            }
+        }
+        parserResultCallable = () -> new ParserResult(
+                new GrobidService(this.preferencesService.getGrobidPreferences()).processReferences(fileList, preferencesService.getImportFormatPreferences())
+        );
+        return parserResultCallable;
+    }
 }

src/main/java/org/jabref/gui/menus/ChangeEntryTypeAction.java

@@ -4,7 +4,6 @@
 
 import javax.swing.undo.UndoManager;
 
-import javafx.beans.property.ReadOnlyStringProperty;
 import javafx.beans.property.ReadOnlyStringWrapper;
 
 import org.jabref.gui.EntryTypeView;
@@ -36,14 +35,4 @@ public void execute() {
                               .ifPresent(change -> compound.addEdit(new UndoableChangeType(change))));
         undoManager.addEdit(compound);
     }
-
-    @Override
-    public String getStatusMessage() {
-        return statusMessage.get();
-    }
-
-    @Override
-    public ReadOnlyStringProperty statusMessageProperty() {
-        return statusMessageProperty.getReadOnlyProperty();
-    }
 }

src/main/java/org/jabref/logic/importer/Importer.java

@@ -168,22 +168,6 @@ public static BufferedReader getReader(InputStream stream) {
         return new BufferedReader(reader);
     }
 
-    /**
-     * Returns the name of this import format.
-     *
-     * <p>The name must be unique.</p>
-     *
-     * @return format name, must be unique and not <code>null</code>
-     */
-    public abstract String getName();
-
-    /**
-     * Returns the type of files that this importer can read
-     *
-     * @return {@link FileType} corresponding to the importer
-     */
-    public abstract FileType getFileType();
-
     /**
      * Returns a one-word ID which identifies this importer. Used for example, to identify the importer when used from
      * the command line.
@@ -202,6 +186,15 @@ public String getId() {
         return result.toString();
     }
 
+    /**
+     * Returns the name of this import format.
+     *
+     * <p>The name must be unique.</p>
+     *
+     * @return format name, must be unique and not <code>null</code>
+     */
+    public abstract String getName();
+
     /**
      * Returns the description of the import format.
      * <p>
@@ -216,6 +209,13 @@ public String getId() {
      */
     public abstract String getDescription();
 
+    /**
+     * Returns the type of files that this importer can read
+     *
+     * @return {@link FileType} corresponding to the importer
+     */
+    public abstract FileType getFileType();
+
     @Override
     public int hashCode() {
         return getName().hashCode();